| dc.description.abstract | Electric load forecasting holds a central position in the domain of power grid management and
sustainable energy development. This study presents an innovative approach that leverages the
integration of external factors within a composite forecasting model that harmoniously merges
Long Short-Term Memory (LSTM) and the Bayesian Optimization (BO) technique. The
research protocol encompassed a comprehensive data collection process, involving the
acquisition of historical load data and the inclusion of a diverse set of external factors, such as
meteorological parameters (MET) and economic indicators (EI). The hybrid model, which
ingeniously combines LSTM with BO, not only demonstrated superior forecasting precision
but also exhibited heightened resilience when compared to conventional forecasting
methodologies. These findings bear compelling implications for optimizing grid management,
resource allocation, and the promotion of persistent energy practices, while also boasting cross industry applicability. This study encourages future research endeavors to further investigate
the integration of additional exogenous factors, thereby refining and extending the hybrid
forecasting model to unlock more intricate insights and enhance the precision of electric load
forecasting, contributing to the continuous advancement in energy resource management and
sustainability. | en_US |
